Better examples of using Intents with Adapt

[KathyReid]

Based on conversations with @penrods, this covers

• how single intents work
• how multiple intents work
• how regex intents work

I'm still not 100% happy with it, but it's better than what we have.

Assigning to @penrods for review (sorry to throw more at you)

[jrwarwick]

great addition! Most uses of "Location" use capital L, but line 446 cites file name with lower case. Is that correct? Does this illustrate some case insensitivity?

Any guidance on how many words (and/or types of words) not represented in the intent elements can appear between represented words? Including but not limited to "unnecessary" adjectives or grammar parts of speech.

Thanks for including reference url at line 432 for re. I think it might yet be helpful to just mention that the regex uses python specific named patterns. Of course someone who cares will figure it out, but you do get the occasional person familiar with regex, but new to python.

[KathyReid]

Thanks for your feedback @jrwarwick, I've added another commit which I think addresses some of your feedback. What do you think?

[KathyReid]

I'm going to merge this for better or worse, I can always do more edits.

[jrwarwick]

Kathy , thanks and sorry for late reply. I think this is definitely an improvement. One thing I'm still not clear about (so maybe just my problem) is how the filename on line 446 relates, or maybe it does not? I believe that with .voc vocab files, the first part of the filename is what appears in the .require() method call argument. But perhaps it is different for regex, and instead of regex filename, the argument represents the Regex group name. So in that case is it correct that when the skill is loaded, all files in regex directory are scanned (filenames don't really matter, just for human understanding) and a list of all regex pattern names is extracted for matching the require() method call's argument. Is that right?
My naive guess/understanding was that require() matched against filenames in both vocab and regex directories rather than file content in one and filename in the other.